Rest Between Sets: How Long, and Why It Varies
By Mark Fulton · 2026-09-03 · 13 min read
For heavy strength and power work, rest 3 to 5 minutes. For hypertrophy, somewhere between 1 and 3 minutes, with about 90 seconds as the point where the returns flatten out. For muscular endurance, under 90 seconds and often 30 to 60. Those ranges are on every fitness site on the internet, and the reason they differ almost never is. Rest length is not a comfort setting. It is the dial that decides how much force you can produce on your next set, and the force you produce across a session is what the strength adaptation is built from. Long rest buys back your output and costs you time. Short rest buys density and costs you reps. Your goal picks which of those two you would rather pay.
The other half of this, and the part that quietly ruins the ranges for most people, is that almost nobody actually times it. "Ninety seconds" is 40 seconds when you feel good and three and a half minutes when your phone comes out. If the interval is the variable, an untimed interval is not a plan.
How long should you rest between sets?
Start with what the research reviews and the position stands actually say, because the numbers on most pages are downstream of these two documents and have been rounded a few times along the way.
The 2009 Sports Medicine review by de Salles and colleagues, Rest interval between sets in strength training, pulled together 35 studies that used rest interval length as the experimental variable. Two of its findings are the backbone of everything since. Acutely, when training with loads between 50 and 90 percent of one-repetition maximum, 3 to 5 minutes of rest between sets allowed for greater repetitions over multiple sets. Chronically, resting 3 to 5 minutes produced greater increases in absolute strength, and the review attributes that to the higher intensities and volumes of training the longer rest made possible. Power showed the same pattern, with higher levels sustained over multiple sets at 3 or 5 minutes of rest versus 1 minute.
The American College of Sports Medicine's position stand on progression models in resistance training for healthy adults turns that into prescriptions. For intermediate and advanced trainees working heavy, meaning loads in the 1 to 6 repetition maximum range, it recommends 3 to 5 minute rest periods between sets. For hypertrophy, with emphasis on the 6 to 12 repetition maximum zone, it recommends 1 to 2 minute rest periods. For power training with light explosive loads it again says 3 to 5 minutes. For local muscular endurance, using light to moderate loads of 40 to 60 percent of 1RM for more than 15 repetitions, it recommends short rest periods of under 90 seconds.
So the popular ranges are real and they are sourced. What gets lost is that they are not four flavours of the same advice. They are four different bets.
Why does the answer change with the goal?
Because rest is doing a different job in each case.
When you finish a hard set, your ability to produce force is temporarily reduced. Rest is the only input that restores it before the next set. The longer you wait, the closer your next set is to a fresh one, which means more reps at the same load, or the same reps at a heavier load. Across a session and then across months, that difference accumulates into total work performed, and total work at a meaningful intensity is the thing strength training is actually trying to accumulate. That is the whole case for long rest, and it is a strong one.
Short rest makes the opposite bet. You accept a worse second and third set in exchange for a session that carries more fatigue per minute. For years the assumption was that this was actively better for building muscle, and the mechanism people cited was hormonal. The 2009 review says it plainly: for hypertrophy, moderate-intensity sets with short rest intervals of 30 to 60 seconds "might be most effective due to greater acute levels of growth hormone during such workouts."
That rationale has not aged well, and it is worth knowing why, because the 30 to 60 second recommendation still circulates on its strength.
In 2016, Schoenfeld and colleagues tested it head on in the Journal of Strength and Conditioning Research. The study, Longer interset rest periods enhance muscle strength and hypertrophy in resistance-trained men, took 21 young resistance-trained men and randomly assigned them to 1 minute or 3 minutes of rest, holding every other training variable constant. Eight weeks, three total-body workouts a week, three sets of 8 to 12 repetition maximum across seven exercises. The 3 minute group came out with significantly greater maximal strength on both squat and bench press, and significantly greater muscle thickness in the anterior thigh, with a trend toward more in the triceps. Muscle endurance improved in both groups with no significant difference between them. Short rest did not win the contest it was supposed to be built for.
The most useful recent read is a 2024 Bayesian meta-analysis in Frontiers in Sports and Active Living, Give it a rest, which pooled 19 measurements from 9 randomised studies comparing inter-set rest durations for muscle growth. Its conclusion is deliberately modest, and it is the sentence to hang your programme on: there appears to be a small hypertrophic benefit to resting longer than 60 seconds, probably because short rest cuts your volume load, but the analysis did not detect appreciable differences in hypertrophy once rest went beyond 90 seconds. In other words there is a floor worth respecting and a ceiling past which extra minutes buy you nothing except a longer session.
The trade-off, by goal
| Goal | Typical rest | What the rest is buying you | What it costs |
|---|---|---|---|
| Max strength (1 to 6 rep range) | 3 to 5 minutes | Near-full force output on every set, so heavy loads stay heavy for the last set as well as the first | Session length. Five sets at five minutes is 25 minutes of standing around, and that is before the warm-up |
| Hypertrophy (6 to 12 rep range) | 1 to 3 minutes, with roughly 90 seconds as the sensible floor | Enough recovery to hold reps and load across sets, which protects the volume that drives growth | Below about 60 seconds you give away reps for a benefit the evidence does not support. Above about 90 seconds you are mostly buying time, not growth |
| Muscular endurance (15+ reps, 40 to 60 percent 1RM) | Under 90 seconds, often 30 to 60 | Accumulated fatigue, which is the specific quality being trained here | Absolute load and total reps at that load. This is not the setting to run when you are trying to add weight to the bar |
| Power (light, explosive, 3 to 5 sets) | 3 to 5 minutes | Speed. Every rep needs to be fast, and a fatigued rep is a slow rep, which is no longer power training | The same time cost as max strength, on a session that looks deceptively easy from the outside |
The column that matters is the last one. Everyone reads the middle column and picks the goal they like the sound of. The cost column is what makes the choice honest, because there is always a cost and it is usually either time or load.
What does short rest actually cost you?
Reps, first and most directly. That is the acute finding from the 2009 review restated: at 50 to 90 percent of 1RM, longer rest allowed more total repetitions across multiple sets. Cut the rest and the drop-off between set one and set three gets steeper. Three sets of eight becomes eight, six and five, and your session volume falls even though the plan on paper did not change.
Second, technique. A set performed on incomplete recovery degrades in a specific way. Bar speed drops, positions loosen, and the last few reps stop looking like the first few. On a leg press that is mostly a wasted rep. Under a loaded bar it is a different conversation.
Third, and least obviously, it costs you a readable record. If your rest varies from set to set and session to session, a good day and a bad day are not comparable, because the rest length changed underneath the numbers you are tracking.
None of that makes short rest a mistake. It is the right tool when total training time is the binding constraint. More sets with brief rests in 40 minutes can beat fewer sets with generous rests in the same 40 minutes, particularly if the goal is conditioning or holding onto muscle rather than adding to a one-rep max. Just choose it as a trade rather than drifting into it.
Does rest length matter more on compound lifts?
In practice yes, though it is worth being precise about why. The position stand does not sort rest by exercise name. It sorts by load and goal, and heavy multi-joint work is simply where the heaviest loads live. A set of five near-maximal squats and a set of twelve cable curls are not two points on the same recovery curve.
There is a second reason that follows from the same logic. Recovery from an isolation set is mostly local to one muscle. Recovery from a heavy squat or deadlift involves your breathing, your trunk, your grip and your general willingness to go again, none of which reset in 60 seconds. The ACSM position stand also recommends sequencing large muscle group exercises before small ones, and multiple-joint exercises before single-joint ones, precisely so the work that needs you fresh gets you first.
A workable default: give the heavy compound work at the front of the session the full 3 to 5 minutes, then shorten toward 90 seconds as you move into accessory and isolation work later on. You lose almost nothing on curls by resting 90 seconds, and you lose a real amount on a top set of squats by resting 90 seconds.
Why is untimed rest always wrong?
Because when you do not time it, something else does, and that something is your current state of comfort.
The pattern is consistent and it runs in two directions at once. Early in the session you feel good, so you go again quickly and the first sets get short rest they did not need. Later in the session you are tired, so the intervals stretch, and the sets that most needed a full recovery get it plus a conversation and a look at your phone. The result is rest that is inversely related to what the training actually called for.
Then there is the measurement problem, which almost nobody raises. Progressive overload depends on comparing this week to last week. If last Tuesday you rested about two minutes, today you rested about four, and today you got an extra rep, you have not learned anything about your strength. You have learned that four minutes beats two minutes, which the research already told you. Untimed rest turns your training log into noise on the exact variable you are trying to read.
Making rest a fixed number solves both problems at once. It does not have to be the perfect number. It has to be the same number.
How do you time rest without babysitting a phone?
The tool needs three things. It should run without an app or an account, it should repeat automatically so you are not resetting it while out of breath, and it should be loud and legible from a few feet away.
The interval timer on this site does that. Set the work length to roughly how long one of your sets takes, set the rest to the interval you picked from the table above, and set the rounds to your number of sets. It flips between a red work phase and a green rest phase with a chime at every switch, and a row of dots shows which set you are on, which is the only thing readable when you are three feet away and breathing hard. Work runs from 5 seconds to 10 minutes and rest from 0 to 5 minutes, so the top of the strength range fits exactly. It holds the screen awake while it runs, and each phase ends at a timestamp taken from your system clock rather than by counting ticks, so a throttled background tab cannot make a round run long. The honest limitation, which applies to every browser timer including this one, is that the tab has to stay open for the sound to fire.
If you would rather have one fixed rest interval that simply repeats all session and follows you around the site, the cooldown timer does that instead. It is set in whole minutes rather than seconds, so it suits a 2 or 3 minute strength rest better than a 90 second one, and it keeps counting while you look at something else.
If your training is interval-shaped rather than set-and-rest shaped, two other formats are worth knowing. Every minute on the minute makes the rest whatever you earn by finishing the work quickly, which is a different and quite useful kind of pressure. And the 20/10 Tabata protocol is what happens when the rest is deliberately too short to recover from, which is the entire point of it and the reason it lasts four minutes rather than twenty.
One plain note before the questions. This is general information about a training variable, not medical or coaching advice. Rest intervals interact with your health, your training history, your medications and whatever you are recovering from, and none of that is visible from a web page. If you have a cardiac or metabolic condition, are pregnant, are returning from injury, or are new to heavy lifting, a clinician or a qualified coach is the right person to ask, and a timer is no substitute for either. It is also worth saying that timing is sometimes not the problem at all. If your sets feel wrong at every rest length, the load, the exercise selection or your sleep is more likely to be the thing to fix than the number on the clock.
Frequently asked questions
Is 60 seconds enough rest between sets?
For moderate-load hypertrophy work it is on the low edge but not useless. The 2024 meta-analysis found a small hypertrophic benefit to resting longer than 60 seconds, likely because shorter rest reduces the reps and load you complete. So 60 seconds is workable when time is tight, and 90 is the better default. For heavy work in the 1 to 6 rep range, 60 seconds is clearly not enough, and both the 2009 review and the ACSM position stand point at 3 to 5 minutes there. For endurance training at light loads and high reps, 60 seconds or less is the intended setting rather than a compromise.
Should rest be longer for heavy compound lifts?
Yes, in the ordinary case. Heavy multi-joint lifts are where the highest loads sit, and the recommendation for those loads is 3 to 5 minutes. The recovery involved is also systemic rather than local, so it takes longer than recovery from an isolation exercise at a moderate load. A practical approach is long rest on the heavy compound work at the start of the session, shortening toward 90 seconds for accessory work later.
Does resting too long reduce the benefit?
Not in the way people usually fear, but the extra minutes stop paying. The 2024 analysis did not detect appreciable differences in muscle growth once rest went past 90 seconds, and the reason it offers is that short rest hurts mainly by cutting volume load, an effect that plateaus once you have rested long enough to keep your reps. So past roughly 90 seconds for hypertrophy, or past 5 minutes for heavy strength work, rest length becomes a session-length decision more than a training one. The real cost of very long rest is that you cool down and the session sprawls, not that you undo anything.
Do supersets change the rest rule?
They change the arithmetic rather than the principle. Pairing two exercises that work opposing or unrelated muscles means each muscle gets the other exercise's working time as extra recovery, so the gap between your own consecutive sets can be longer than it looks even though you never stop moving. Judge the rest per muscle rather than per minute on the clock, and hold the same target you would have used for those exercises separately. Worth noting that the evidence discussed above tested straight sets with a controlled interval, so treat superset timing as a sensible extension of the same reasoning rather than something those studies measured. And if you are pairing two heavy compound lifts, be honest about what that is: a conditioning choice, and one that will cost you load.